memory_append

You can persist project context notes between sessions by appending content to a memory file. This skill automates appending notes to the .kogniterm/llm_context.md file, ensuring contextual data remains available across different working sessions.

Yes, you can append notes to a custom memory file by providing an optional file_path parameter. This overrides the default llm_context.md target, enabling you to maintain multiple separate memory contexts for different project aspects.

The best way to automate adding contextual notes is using a simple append operation that writes in UTF-8 encoding. This skill safely targets the memory store, creating the .kogniterm directory if it is missing, and appends your content directly.

Yes, you need the memory permission to write to the llm_context.md memory file. This permission allows the skill to create the .kogniterm directory if it does not exist and perform the append operation safely.

Project context disappears because it is not persisted by default. Appending important notes to a memory file like llm_context.md solves this by storing the contextual data in your project directory, preserving it for future sessions.

A limitation of using a simple append operation is that it only adds content to the end of the memory file. It cannot modify or organize existing notes, so the file grows sequentially without structured formatting or automated cleanup.